Surfing the Cosmic Reef

This blog post is one in a series marking the 30th anniversary of the launch of the Hubble Space Telescope. For more information and resources regarding Hubble’s 30th anniversary, please visit hubblesite.

Thirty years is a long time. In fact, thirty years is roughly the span of a generation. Thus, it is no exaggeration to say that an entire generation of people have had their perception of the universe shaped by the Hubble Space Telescope’s keen vision. Three decades in orbit and still going strong, Hubble continues to provide us with spectacular cosmic vistas (and take a look at Ray Villard’s recent post for more on Hubble’s history). To mark this significant milestone in Hubble’s journey, we wanted to produce another iconic view of the universe that only Hubble could provide.

The Hubble Space Telescope photographed by the astronauts of Servicing Mission 3B

Before he retired in 2018, Hubble image guru Zolt Levay had investigated several potential targets suitable for the thirtieth anniversary. Out of curiosity and a fascination with numerology, he looked into potential numerical coincidences in the New General Catalogue (NGC). Naturally, with the anniversary falling in the year 2020, he looked at NGC 2020, located just outside the Milky Way in a satellite galaxy known as the Large Magellanic Cloud, or LMC for short. The LMC sits about 160,000 light-years from us, and contains some of the most enigmatic regions in our local universe. It turned out to be a very compelling object and ultimately became our target of choice to commemorate Hubble’s thirty years in orbit. Max Mutchler has written a blog post detailing our decision-making process and observing strategy for this object, so I refer you to that post for more details. Here, we’ll take a closer look at what it is, and how we assembled the final image.

What is NGC 2020?

As you gaze into this image, you will likely notice two very distinct objects. NGC 2020 is the smaller, circular blue nebula in the lower left; it is actually a Wolf-Rayet bubble. The region that dominates most of the image is a star cluster and star-forming nebula known as NGC 2014. Wolf-Rayet (WR) stars are incredibly hot with very strong, high-energy ultraviolet stellar winds. A WR nebula forms when these strong stellar winds interact with layers of gas that have been ejected by the star. Look closely near the center of NGC 2020 and you’ll notice a bright star, known as Br 48 (or HD 269748), that is the WR star generating this nebula.

While the general shape and appearance of NGC 2020 has been determined by a single star, NGC 2014’s delicate tendrils and loosely packed bubbles (resembling a brain coral) have all evolved from the interactions of stellar winds generated by the bright cluster of young, blue stars in the center of the nebula. The winds from these stars have actually blown bubbles into the gas and dust in the surrounding nebula, carving the ethereal beauty captured in exquisite detail by Hubble. The densest region within NGC 2014 sits right at the center of our image. Here, like turquoise waves washing up on the shore, the combined winds of the star cluster are colliding with dense gas, causing it to glow brightly in visible light in a process known as ionization.

The brightly glowing ionization fronts of NGC 2014

Within this bright region, beyond the shore, you’ll also notice a mountainous landscape of gas and dust not unlike those that Hubble has spotted in our own galaxy, like the famous Pillars of Creation in the Eagle Nebula. These towers of gas are denser than their surroundings, allowing them to resist erosion by the barrage of stellar winds, and carving them into their spire-like shapes.

The Pillars of Creation

Viewing this image at its native resolution is highly recommended. Drilling down from the wide-field view, there’s a cornucopia of interesting features to be discovered by the wandering, curious eye. From the bright cave within NGC 2014 to the independent mini-nebulas scattered around the edges, there’s even a background galaxy hiding in plain sight, shining through the gas and dust of NGC 2014. As we’ve seen time and again with Hubble, there’s far more to the universe than meets the eye.

NGC 2014’s “bright cave” and a background galaxy

A Tsunami of Pixels

As Max details in his post, our main goals in devising an observing strategy for this object were twofold: the first was to maximize the size of the mosaic giving us the largest image possible within our observing constraints; the second was to strike a delicate balance between capturing enough light in our chosen filters to see faint structures within the nebulas, while not over-saturating the very bright stars within the central cluster. There were many iterations of observing plans to reach these goals, and as the data started coming down in early January, we were extremely pleased with the results.

Our first look at NGC 2020

Of course, even with the best-laid plans, circumstances beyond your control can throw a wrench into everything. While hurtling through orbit around the earth at 18,000 miles per hour, Hubble has to maintain its precise pointing in order for an observation to produce meaningful data. It does this through the use of guide stars; bright stars in the field of view that Hubble can lock onto and track throughout the observation. On rare occasion, Hubble may lose track of those guide stars and begin drifting, smearing out the resulting image. This occurred in three out of the twelve tiles of our mosaic causing us to lose those data, and leaving a gaping hole in the middle of our beautiful image!

A first draft of the mosaic with missing tiles

Thankfully, we were eventually able to revisit those observations and successfully fill in the remainder of the image although with a slight twist, literally! The orientation of Hubble’s detectors on the sky, or its ‘nominal roll angle,’ changes through time as Hubble follows earth’s orbit around the Sun. The makeup observations were scheduled as soon as possible, yet about six weeks had elapsed before Hubble could look at NGC 2020 again. This meant that the resulting data were rotated with respect to the original tiles and made for some interesting triangular artifacts in the data which were mostly handled through calibration when merging the data together. The image below has been heavily stretched to emphasize those artifacts which were later cleaned up.

A single mosaic image shown in grayscale to illustrate the rotated tiles and background noise

The narrow-band filters used in these observations were specifically chosen to highlight the light of different gasses commonly found in star-forming nebulas, including oxygen and hydrogen. We also used two very broad filters to collect the blue and red light of stars within the field and cluster, allowing us to compose an image that shows the stars in natural colors balanced with the delicate features of the nebula. Following our usual convention of assigning colors chromatically according to wavelength, the longest wavelength filter, the broad red light filter, is colored red, moving through orange, teal, and blue with shorter wavelengths. However, due to the different amounts of light transmitted by each filter, color-balancing the image was a challenge.

The final full mosaic with filter and color assignments

Waves of Color

As the data streamed down from Hubble, our first few preliminary images were composed using only the narrowband data as two-color images, simply based on what data were available at the time. This produced striking images of the nebula, but left the star colors mostly washed out with little color definition.

One of our first views of NGC 2020 in two-color mode
Initial views of the Brain Coral in NGC 2014 in two-color mode
Assembling the mosaic in two-color mode

We know that the bright stars in the cluster at the center of the nebula are mostly blue, yet they were not appearing blue in the two-color images, clearly illustrating the need for all four filters for a balanced image. The broad-band data consisted of very short exposures to avoid saturating the bright stars. As a result, there was very little nebula captured in the broad-band images and the data also exhibited a significant level of background noise. This was of course the result of the tradeoffs we made in our observation planning, given our constraints.

The final mosaic seen in individual filters, along with the initial composite before any color balancing, and the final composite

A careful stretching of the broadband data was required to minimize the amount of noise introduced into the image. The overall color of the composite image was very sensitive to the initial stretch of all four individual images, requiring a meticulous approach to the color balance of the image to ensure that NGC 2020 appears blue, while NGC 2014 shows off its various shades of hydrogen-dominated red emission, all while maintaining the natural color of the cluster stars. After many iterations, we had this whole process roughly mapped out before the final gap-filling frames came down. Obviously, we knew there would be more nebula contained in that missing tile, but the level of detail revealed when that final central panel was included, completing the image, was a breathtaking moment.  We had a strong hunch that NGC 2020 and NGC 2014 were going to make a striking pair and a spectacular image from the start. Yet, even us seasoned veterans of Hubble imagery were awestruck by the sheer beauty of this little corner of the universe.

After thirty years, Hubble is now in somewhat uncharted waters. This is the longest the telescope has ever gone in its lifetime without a servicing mission, but it is a workhorse observatory, continually churning out groundbreaking science and seeking answers to humanity’s most fundamental questions about our place in the universe. It is very fortunate for all of us that in the process of seeking those answers, Hubble also provides us with extremely detailed glimpses into the unseen beauty surrounding us in the universe. I can’t wait to see what Hubble will show us next!

3 thoughts on “Surfing the Cosmic Reef

Add yours

  1. Do you ever have a difficulty compensating for anomalies when combining x-ray imagery and imagery from radio telescopes? Do the Stars and phenomena have a tendency not to align precisely? For example in this amazing image, would x-ray Imaging and otherwise invisible sources fail to align with purely visible or radio telemetry? Thank you for entertaining my questions. I’d like to learn more about that correction if it is required. Also I want to congratulate you on this phenomenal image and I love hearing how you made so much of it boss available and understandable.

    Like

    1. Yes, combining data from different telescopes can be a daunting task requiring great attention to detail and accuracy. The first hurdle to jump through is aligning the field of view and angular resolution of different telescopes. The images all have to be scaled so that when they overlap, we can be sure that all images are showing the exact same portion of the sky oriented in exactly the same way. Thankfully, most astronomical data comes with the coordinate information of the image embedded in the data and astronomical analysis software is very adept at translating coordinates from one image to another. Once the images are aligned, we frequently see that there are features in one wavelength, like x-ray that do not overlap with anything in the optical image. For example, background AGN or active galactic nuclei are very bright in x-rays, but may not show up at all in optical images. Thanks for reading!

      Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create a website or blog at WordPress.com

Up ↑

%d bloggers like this: